Extract links from specific table

Question

I have a html code with many html tables. I want to extract links from specific one which has specific div above.

Here's my sample code:

        <div class="boxuniwersal_header">Table 1</div>
        <img src="img/boxuniwersal_top.gif" width="210" height="18" alt="" style="margin-top: 5px" />
        <div class="boxuniwersal_content">
          <div class="boxuniwersal_subcontent">
            <div class='menu_m1'><table cellpadding="3"><tr><td><a href="http://link.com"><img src="some.jpg"  width="45" /></a></td><td><a href="http://link.com">Some text</a></td></tr></table></div>
            <br />
          </div>
        </div>
        <!-- /box  -->

        <!-- box  -->
        <div class="boxuniwersal_header">Table 2</div>
        <img src="img/boxuniwersal_top.gif" width="210" height="18" alt="" style="margin-top: 5px" />
        <div class="boxuniwersal_content">
          <div class="boxuniwersal_subcontent">
            <div class='menu_m1'><table cellpadding="3"><tr><td><a href="http://link2.com"><img src="some2.jpg"  width="45" /></a></td><td><a href="http://link2.com">Some text2</a></td></tr></table></div>
            <br />
          </div>
        </div>

$domXPath = new DOMXPath($domDocument);
$results = $domXPath->query("//div/div/table/tr/td/a|//table//tr/td//a"); //querying domdocument
foreach($results as $result)
{
    $links[]=$result->getAttribute("href");
}

This code returns all links. I want to grab only links from Table1. Is it possible?

Answer 1

Your main problem is just tuning the XPath expression to select the right XML.

If you change your XPath to

//div[text()="Table 1"]/following-sibling::div[1]//table//a

What this does is first find the <div> element whose text is the one your after.

The following-sibling::div[1] part will look at the first <div> element at the same level as the <div> element already selected (this is the one where the <table> is).

The last part just looks for all <a> elements within the enclosing <table>.